INDEX

Explanations

forms of the word "cum" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

/stdc

-0.16

vest

-0.16

 Surround

-0.15

entlich

-0.14

lash

-0.14

pend

-0.14

agers

-0.14

erts

-0.14

ç§°

-0.14

boro

-0.14

POSITIVE LOGITS

teb

0.16

tez

0.15

ullan

0.14

roperties

0.14

 Heads

0.14

 typename

0.14

 Basin

0.13

eyJ

0.13

 scho

0.13

Probe

0.13

Activations Density 0.012%

forms of the word "cum" in various contexts

No Comments

No Known Activations